The purpose of this study is to investigate by sonography the impact of intracatheter nitroglycerin infusion upon the incidence of vascular thrombosis following percutaneous placement of femoral venous catheters in critically ill children. We hypothesize that intracatheter nitroglycerin infusion will be associated with a lower incidence of thrombosis when compared to control catheters. The study will include a total of 62 children ages birth to 6 years of age. Two groups will be studied. Group 1 (nitroglycerin; n=31) will include critically ill children admitted to the intensive care unit in whom the primary care team has elected to place a double or triple lumen femoral venous catheter. Group 2 (D5W; n=31) will also include critically ill children admitted to the intensive care unit in whom the primary care team has elected to place a double or triple lumen femoral venous catheter. Ultrasound examinations of the catheterized femoral vein will be performed and analyzed by the Radiology investigator within 2 days of catheter insertion and a final study will be performed prior to discharge from the hospital. The primary outcome variable is the difference in the incidence of ultrasound-diagnosed thrombosis between the nitroglycerin and D5W groups. Secondary outcome variables include difference in clinical evidence of thrombosis between the two groups, difference in the incidence of catheter-related infection between the two groups, impact of TPN, intralipid, and heparin infusions on thrombosis in each group, and impact of catheter duration on thrombosis in each group. To date, 40 of the 62 needed participants have been enrolled. Enrollment has been a limiting factor, but if we can get enrollment up to an average of 2 per month, the study could be completed in 12 months. Enrollment can be increased with more aggressive education of the fellows and increased availability of the study coordinator. No interim results are available at this time.